PARIS – The 5-year mortality rate associated with untreated moderate aortic stenosis is just as grim as it is for severe aortic stenosis, according to new findings from the largest-ever study of the natural history of aortic stenosis.
Dr. Geoff Strange
“These data provide a clear signal of the expected adverse outcomes for individuals presenting across the globe with a mean aortic valve gradient greater than 20.0 mm Hg or a peak aortic valve velocity above 3.0 m/sec,” Geoff Strange, PhD, said in presenting an analysis from NEDA, the National Echocardiography Database of Australia, at the annual congress of the European Society of Cardiology.
These results, if confirmed in other large datasets, could potentially have enormous implications for the use of transcatheter and surgical aortic valve replacement, interventions which until now have been restricted to patients with severe aortic stenosis (AS) as defined by an aortic valve (AV) mean gradient in excess of 40 mm Hg or a peak AV velocity greater than 4.0 m/sec. This restriction was based on what Dr. Strange considers rather limited and flimsy evidence suggesting that the mortality associated with AS was negligible except in severe AS.
“Cut points used to stratify for interventional strategies are based on very small numbers,” observed Dr. Strange, professor of medicine at the University of Notre Dame in Fremantle, Australia.
The NEDA findings, he added, constitute a call to action: “These data provide the impetus for a contemporary evaluation of the risk-to-benefit ratio of intervention in the moderate AS population,” Dr. Strange declared.
He and his NEDA coinvestigators analyzed echocardiographic data on 241,303 individuals in the Australian database, zeroing in on the 25,827 with untreated mild, moderate, or severe native valve AS. To place the size and scope of this project into perspective, the next-largest study of the natural history of untreated AS included 1,375 individuals – and that study was in turn roughly 10-fold bigger than the handful of other published studies addressing this issue.
A key finding in the NEDA study was that the 5-year all-cause mortality rate of 61.4% in the group with moderate AS wasn’t significantly different from the 64.6% rate in those with severe AS (see graphic).
The investigators performed additional analyses, analyzing peak velocity and mean gradient as continuous variables and stratifying patients into quintiles on that basis. They found that the top quintile for AV velocity started very low, at 1.73 m/sec, while the top quintile for mean AV gradient also started at a surprisingly low level: greater than 9.6 mm Hg. They noted that both all-cause and cardiovascular-specific mortality rates were basically flat until taking what Dr. Strange described as “a sharp pivot point upward” right around 20 mm Hg or 3 m/sec.
“No matter how we looked at these data – whether we looked at patients with or without left heart disease, whether we used the dimensionless index, whether we adjusted for stroke volume index, whether we stratified between age above or below 65, whether we used the gradient, the velocity, or the AV area – this threshold of increasing mortality at around 20 mm Hg or 3 m/sec continued to emerge,” according to Dr. Strange.
He noted that this study used real-world data with hard endpoints – actuarial patient mortality outcomes obtained through linkage to the national database – rather than hypothetical projections based upon Kaplan-Meier curves. A study limitation was that comorbidity data couldn’t be obtained for the AS patients.
Session cochair Patrizio Lancellotti, MD, PhD, commented, “I think this study will change a bit our consideration about patients with moderate AS.”
However, Dr. Lancellotti, who was lead author of the second-largest study of the natural history of aortic stenosis (JAMA Cardiol. 2018 Nov 1;3[11]:1060-8), expressed misgivings about the NEDA system’s lack of a core echocardiographic laboratory for imaging adjudication. That’s a study weakness given that image quality and the accuracy of echocardiographic interpretation are so highly dependent upon an individual cardiologist’s skill, observed Dr. Lancellotti, who is head of cardiology at the University of Liege (Belgium).
Dr. Strange replied that he and his coinvestigators analyzed a random subset of the NEDA data and found very little interlaboratory variability in results.
“All I can say is that the labs that contributed to this study are the most eminent labs across Australia,” he added.
Simultaneously with Dr. Strange’s presentation at the congress, the NEDA study results were published online (J Am Coll Cardiol. Sep 2019. doi: 10.1016/j.jacc.2019.08.004).
Dr. Strange reported having no financial conflicts of interest regarding the NEDA project, which is funded by GlaxoSmithKline, Bayer, and Actelion.